snorkel.classification.LogManager

class snorkel.classification.LogManager(n_batches_per_epoch, log_writer=None, checkpointer=None, **kwargs)

Bases: object

A class to manage logging during training progress.

Parameters

• n_batches_per_epoch (int) – Total number batches per epoch

• log_writer (Optional[LogWriter]) – LogWriter for current run logs

• checkpointer (Optional[Checkpointer]) – Checkpointer for current model

• kwargs (Any) – Settings to update in LogManagerConfig

__init__(n_batches_per_epoch, log_writer=None, checkpointer=None, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

Return type

None

Methods

__init__(n_batches_per_epoch[, log_writer, …])	Initialize self.
cleanup(model)	Close the log writer and checkpointer if needed.
reset()	Reset counters.
trigger_checkpointing()	Check if current counts trigger checkpointing.
trigger_evaluation()	Check if current counts trigger evaluation.
update(batch_size)	Update the count and total number.

cleanup(model)

Close the log writer and checkpointer if needed. Reload best model.

Return type

MultitaskClassifier

reset()

Reset counters.

Return type

None

trigger_checkpointing()

Check if current counts trigger checkpointing.

Return type

bool

trigger_evaluation()

Check if current counts trigger evaluation.

Return type

bool

update(batch_size)

Update the count and total number.

Return type

None